Text Summarization Using PEGASUS Transformer Model in Machine Learning

Extracting crucial information from lengthy documents can be a time-consuming and labor-intensive process. Automatic text summarization algorithms address this challenge by condensing extensive texts while retaining essential details. The development of efficient summarization systems is particularly important for complex languages. This study explores the application of Transformer-based Language Models (TLMs) in multi-dataset text summarization. It compares the effectiveness of deep learning and traditional machine learning techniques against TLMs. The findings indicate that TLMs, specifically PEGASUS models, outperform baseline methods, achieving an average F1-Score of $\mathbf{9 0 \%}$ across a diverse range of datasets.